Imagery Tools for Small-Plot Research - Deveron UAS · observations by drones, researchers can improve the power of studies by redeploying resources to conducting trials in more ...

1

Imagery Tools for Small-Plot Research

2

Executive Summary

Crop researchers face increasing pressure to move genetics,products, and management practices through the selectionprocess. Drone data is a solution to the labour and time constraintsthat limit the number of data points that can be collected in asingle season. Using the rapid and simultaneous collection of plotobservations by drones, researchers can improve the power ofstudies by redeploying resources to conducting trials in moreenvironments and with greater depth.

Deveron’s research offerings treat imagery like any other plotrecord so that it integrates with typical statistical analyses. Ourproducts include plant counting, vegetative fraction, crop height,canola flowering, and multispectral analysis. Our expertise inremote sensing and crop science can be relied upon to furtherextract insight from imagery according to the specific needs of theresearcher.

3

Who is Deveron?

Ohio Indiana Illinois Iowa Missouri Kansas Nebraska South Dakota North Dakota Minnesota Expected growth into

southern states this winter

Alberta Saskatchewan Manitoba Ontario Quebec

Deveron is a full-service drone imagery provider operating a network of pilots across Canada and the US. In 2018, we were active in the following regions:

4

Deveron’s Drone Data Service Platform

Order and download products using our SOAR platform

• Receive email updates when flights are logged and completed

• Upload trial boundaries via shapefile draw directly in geospatial app

5

Why drones for research?

Research is a numbers-limitedendeavour. The power of a study is limited by:

• Treatment count

• Environmental replication

• Available labour

There is always a trade-off when designing experimentsMore trials, extensive study

OR

Fewer trials, intensive study

6

A Picture Says a Thousand Numbers Observation

Information

To be useful in statistical analyses, imagery must be transformed from visual observation into quantified information.

Each of Deveron’s services include the identification and extraction of plot-level data into spreadsheets to be used in treatment analyses.

We also include a geospatial shapefile of plots to include with field records, and for reanalyzing trials according to new insights gained over time.

7

Drone Data Services for Researchers

7

8

Basic NDVI/NDRE• General indicator of

plant stress• Plot greenness for early-

to mid-season (NDVI) and late-season (NDRE)

• Index maps compatible with GIS and imagery softwares

5-band Reflectance• Full RGB + NIR + red

edge coverage• Plot means for

calculating any index in spreadsheet

• Reflectance Maps compatible with GIS and imagery softwares

Vegetative fraction (% green pixels)

• Estimate canopy extent using green pixels

• Plant-only mask enhances

Data Services

9

Data Services

Canola Flowering• Estimate flowering

extent from yellowness index

Plant Count• 5mm resolution RGB• Identify individual corn

plants up to V4 (leaf-tip) • Other crops with large

plant-to-plant separation

Plant Height• 3D model of plot• Estimate height using

bareground map• Combine with

vegetative fraction and multispectral for crop modelling

10

Imagery Example: Winter Wheat Variety Trial

Winter wheat NDVI Winter wheat RGB

11

Plot Data Example: Factorial Corn Trial

Grain corn with 4x replicated experimental design

12

Vegetative Fraction Example: Soybean Senescence

Soybean yield trial. Pre-canopy (left), pre-harvest interval (centre), and extent of pre-harvest leaf drop (right).

15

For additional information please contact

Jacob NederendResearch Agronomist—Deveron UAS

[email protected]

15